1288 volume 44 | number 12 | december 2012 | nature genetics
independent studies in this issue of Nature Genetics by Richter et al.5 and by Love et al.6 as well as a recent study in Nature by Schmidt et al.7.
Burkitt lymphoma is an aggressive B-cell neo-plasm initially recognized by Dennis Burkitt, a prescient surgeon working in Uganda in the mid-dle years of the last century1. With only a handful of collaborators, he was able to characterize an obscure tumor affecting the jaws and abdomi-nal organs of children. These tumors were later observed in Europe and North America and, in contrast to the endemic distribution in central Africa, they were considered sporadic BL. The same tumor was also frequently recognized in patients with acquired immunodeficiencies1. The identification of BL opened the pathway toward the first recognition of two important findings for understanding the pathogeneiss of different types of tumors: the Epstein-Barr virus (EBV) and the MYC oncogene, target of the t(8:14) translocation present in virtually all BL1,2. Although EBV and the MYC translocation have been considered to be important players in BL pathogenesis, neither is sufficient to explain the behavior of the tumor3. The high proliferation of BL could be related to dysregulation of the cell cycle induced by the MYC protein, but MYC also activates apoptosis, and therefore BL cells must have mechanisms to overcome this tumor-protective effect3. The t(8:14) translocation has also been detected in blood cells of healthy individuals, reinforcing the idea that it is not sufficient to drive tumor development4. On the other hand, although EBV is present in the vast majority of endemic and immunodeficiency-associated BL, it is detected in only 30% of sporadic cases of BL3.
New somatic mutationsThese unresolved questions about the patho-genesis of BL have found initial answers in two
new pathogenic mechanisms in Burkitt lymphomaElias Campo
Two studies in this issue identify the landscape of somatic mutations in Burkitt lymphoma and highlight the pathogenic and clinical relevance of inactivating mutations of ID3, an inhibitor of the TCF3 transcription factor.
Elias Campo is at the Hospital Clinic, Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), University of Barcelona, Barcelona, Spain. e-mail: [email protected]
Using integrative structural and functional genomics, these studies have provided the first catalogue of somatic mutations in BL, thus
200 µm
50 µm
50 µm
LZ
DZ
a b
Germinalcenter
function
ProliferationSurvival
CCND3
PI3K
TCF3 ID3
MYCBCR
ProliferationSurvival
CCND3
PI3K
BCR
Cellgrowth
Germinalcenter
function
TCF3 ID3
MYC
Cellgrowth
Figure 1 Somatic mutations in Burkitt lymphoma frequently target ID3, TCF3 and CCND3. (a) TCF3 modulates germinal center genes and is preferentially expressed in the highly proliferative area of this structure, recognized histologically as the dark zone (DZ). TCF3 also regulates survival and proliferation of lymphoid cells through the BCR and PI3K signaling pathway and by modulating cell cycle regulators such as CCND3. TCF3 also induces its own inhibitor ID3, creating an autoregulatory loop that may attenuate this program and facilitate the transition of the germinal center cells to the light zone (LZ). MYC is not normally expressed in cells of the DZ and is upregulated in the LZ. Its induction of ID3 may contribute to the attenuation of the TCF3 pathway in the normal germinal center. (b) Burkitt lymphoma (top) frequently harbors mutations in ID3, TCF3 and CCND3 that activate the TCF3 pathway. The t(8:14) translocation present in BL dysregulates MYC. Cooperation of these two pathways plays a crucial role in BL, in which virtually all cells are proliferating, as evidenced by the expression of the cell cycle– related antigen Ki67 (bottom).
nature genetics | volume 44 | number 12 | december 2012 1289
identifying new oncogenic pathways and mutated genes, some of which were virtually unknown in cancer. These findings also have clinical relevance, providing elements for refined diagnosis and new therapeutic strategies.
The most surprising finding was the frequent somatic mutations in the genes encoding the transcription factor TCF3 and its negative inhibitor ID3, affecting up to 70% of sporadic and immunodeficiency-associated BL and 40% of endemic tumors. ID3 mutations (38–68%) were more common than TCF3 mutations (11%), and most of the former were inactivating and biallelic or were associated with deletion of the other allele, suggestive of a tumor- suppressor function. In contrast, TCF3 muta-tions were monoallelic and occurred at conserved residues, suggesting that they could be activating. Functional studies showed that action of the TCF3 protein and mutated ID3 were critical for the survival and proliferation of BL cells7. Schmidt et al. studied the transcriptional program of TCF3 in BL cells and found that it modulates essential genes for germinal center function (Fig. 1a) and also upregulates ID3, creating an inhibitory loop that in normal cells would attenuate TCF3 action7. Interestingly, the ID3 locus is a direct target of MYC and may also contribute to downregulating the TCF3 path-way in normal cells8. Therefore, it is likely that the inactivating mutations of ID3 in BL release TCF3 from its inhibitory function (Fig. 1b). TCF3 also promoted the survival of BL cells by intensifying B-cell receptor (BCR) signal-ing through the phosphoinositide-3-kinase (PI3K) pathway, and it promoted their prolif-eration by modulating cell cycle–related genes such as CCND3 (ref. 7) (Fig. 1a). Interestingly, activating mutations in CCND3 were also found
in 38% of sporadic BL but only in a minority of endemic tumors5,7. These results are remark-able because a recent mouse model in which PI3K signaling cooperates with MYC develops a lymphoma that recapitulates the characteris-tics of human BL, including frequent CCND3 mutations9. These observations strongly sup-port the cooperation of MYC translocation with the mutations of ID3, TCF3 and CCND3 in the pathogenesis of BL (Fig. 1b).
Richter et al.5 concluded that ID3 mutations have the imprint of the mutational machinery of the germinal center, a mechanism that acts physiologically on the immunoglobulin genes to increase their affinity for the respective antigen5. This finding is intriguing because diffuse large B-cell lymphomas (DLBCL), a different type of aggressive lymphoma that also originates in cells that have experienced this mutational microenvi-ronment, do not have ID3 mutations5–7. The rea-son for this apparent discordance is not clear, but it suggests that BL and DLBCL may originate in cells of the germinal center with different depen-dence of the ID3-TCF3 pathway. The lymphoid germinal center has two distinct topographic and functional areas, recognized as the dark and light zones (DZ and LZ) (Fig. 1a). The DZ is highly proliferative, whereas cells in the LZ are begin-ning to differentiate towards effector cells. TCF3 is more abundant in DZ than in LZ cells, and the expression profile of BL is more similar to that of DZ cells whereas the profile of DLBCL is closer to that of LZ cells, supporting the idea that these two types of lymphomas may arise from different subpopulations of the germinal center10.
Clinical implications and next stepsThe finding of ID3 mutations exclusively in BL may help in the diagnosis of aggressive
lymphomas11. Richter et al. found that tumors with a gene expression profile intermediate between those of BL and DLBCL but carry-ing ID3 mutations had clinical and biological features closer to those of conventional BL5. Current therapeutic protocols have improved the outcome in children but are less successful in adults12, although the toxicity of these treat-ments is high and they are difficult to apply in low-income countries3,12. The identifica-tion of this new pathogenic pathway essen-tial for BL cells may provide new potential therapeutic targets.
The mutations recognized in these studies do not seem to be present in all patients, and their different distribution in sporadic and endemic BL is intriguing7. It will be important to clarify whether the remaining patients have mutations in genes of the same or other pathways. The relationship between these genetic alterations and EBV infection also needs further study, particularly in relation to the different distribu-tion of the virus and ID3 or CCND3 mutations in the epidemiological subtypes of the disease.
COMPETING FINANCIAL INTERESTSThe author declares no competing financial interests.
1. Wright, D. Br. J. Haematol. 156, 780–782 (2012).2. Dalla-Favera, R. et al. Proc. Natl. Acad. Sci. USA 79,
7824–7827 (1982).3. Magrath, I. Br. J. Haematol. 156, 744–756 (2012).4. Janz, S. et al. Genes Chrom. Cancer 36, 211–223
(2003).5. Richter, J. et al. Nat. Genet. 44, 1316–1320 (2012).6. Love, C. et al. Nat. Genet. 44, 1321–1325 (2012).7. Schmitz, R. et al. Nature 490, 116–120 (2012).8. Seitz, V. et al. PLoS One 6, e26837 (2011).9. Sander, S. Cancer Cell 22, 167–179 (2012).10. Victora, G.D. et al. Blood 120, 2240–2248 (2012).11. Salaverria, I. & Siebert, R. J. Clin. Oncol. 29,
1835–1843 (2011).12. Molyneux, E.M. et al. Lancet 379, 1234–1244
(2012).
focus on errors of RNA metabolism and processing. The catalyst for this shift was the discovery that 43-kDa TAR DNA-binding protein (TDP-43) is a major component of ubiquitinated protein aggregates found in sporadic amyotrophic lateral sclerosis (ALS) and the most common form of frontotem-poral dementia (disease with ubiquitinated inclusions, or FTLD-U)1,2. Subsequently, mutations in the genes encoding TDP-43
A paradigm shift in understanding the mechanisms of several neurodegenerative diseases is underway, with an increasing
TdP-43 toxicity and the usefulness of junkShuying Sun & Don W Cleveland
a new study shows that loss of the lariat debranching enzyme dbr1 suppresses TdP-43 toxicity. The accumulated intronic lariat Rnas, which are normally degraded after splicing, likely act as decoys to sequester TdP-43 away from binding to and disrupting functions of other Rnas.
Shuying Sun and Don W. Cleveland are at the Ludwig Institute for Cancer Research and Department of Cellular and Molecular Medicine, University of California at San Diego, La Jolla, California, USA. e-mail: [email protected]
and a related RNA binding protein, FUS/TLS, were found to cause both diseases. Moreover, TDP-43 immunoreactive inclusions have been observed not just in inherited and sporadic ALS and FTLD-U, but also in Alzheimer’s, Parkinson’s and Huntington’s diseases3. Typical pathology in affected individuals includes reduced accumulation of TDP-43 in nuclei and the presence of cytoplasmic inclusions4. While the disease mechanisms
